Complexity of cyanobacterial exopolysaccharides: composition, structures, inducing factors and putative genes involved in their biosynthesis and assembly


  • Sara Pereira,

    1. IBMC – Instituto de Biologia Molecular e Celular, Universidade do Porto, Porto, Portugal
    2. Departamento de Botânica, Faculdade de Ciências, Universidade do Porto, Porto, Portugal
    Search for more papers by this author
  • Andrea Zille,

    1. IBMC – Instituto de Biologia Molecular e Celular, Universidade do Porto, Porto, Portugal
    Search for more papers by this author
  • Ernesto Micheletti,

    1. Department of Agricultural Biotechnology, University of Florence, Florence, Italy
    Search for more papers by this author
  • Pedro Moradas-Ferreira,

    1. IBMC – Instituto de Biologia Molecular e Celular, Universidade do Porto, Porto, Portugal
    2. Instituto de Ciências Biomédicas Abel Salazar (ICBAS), Universidade do Porto, Porto, Portugal
    Search for more papers by this author
  • Roberto De Philippis,

    1. Department of Agricultural Biotechnology, University of Florence, Florence, Italy
    Search for more papers by this author
  • Paula Tamagnini

    1. IBMC – Instituto de Biologia Molecular e Celular, Universidade do Porto, Porto, Portugal
    2. Departamento de Botânica, Faculdade de Ciências, Universidade do Porto, Porto, Portugal
    Search for more papers by this author

  • Editor: Ferran Garcia-Pichel

Correspondence: Paula Tamagnini, IBMC – Instituto de Biologia Molecular e Celular, Universidade do Porto, Rua do Campo Alegre, 823, 4150-180 Porto, Portugal. Tel.: +35 122 607 4900; fax: +35 122 609 9157; e-mail:


Cyanobacterial extracellular polymeric substances (EPS) are mainly composed of high-molecular-mass heteropolysaccharides, with variable composition and roles according to the microorganism and the environmental conditions. The number of constituents – both saccharidic and nonsaccharidic – and the complexity of structures give rise to speculations on how intricate their biosynthetic pathways could be, and how many genes may be involved in their production. However, little is known regarding the cyanobacterial EPS biosynthetic pathways and regulating factors. This review organizes available information on cyanobacterial EPS, including their composition, function and factors affecting their synthesis, and from the in silico analysis of available cyanobacterial genome sequences, proposes a putative mechanism for their biosynthesis.


A wide range of bacteria are able to synthesize and secrete extracellular polymeric substances (EPS) (Neu & Marshall, 1990), mainly of polysaccharidic nature, which can remain covalently linked or loosely attached to the cell surface, or be released into the surrounding environment (De Philippis & Vincenzini, 2003). These exopolysaccharides can be divided into two groups: homopolysaccharides and heteropolysaccharides (Sutherland, 2001). The homopolysaccharides are composed of only one type of monosaccharide, and are synthesized from sucrose by the action of a sucrase. The heteropolysaccharides consist of high-molecular-mass hydrated molecules made up of different sugar residues, and are synthesized by the combined action of different types of glycosyltransferases (De Vuyst & Degeest, 1999; De Vuyst et al., 2001; Van Hijum et al., 2006; Arskold et al., 2007). These complex polymers can also contain acetylated amino sugar moieties, as well as noncarbohydrate constituents such as phosphate, lactate, acetate and glycerol (De Vuyst & Degeest, 1999; De Vuyst et al., 2001; Ruas-Madiedo et al., 2002; Girard & Schaffer-Lequart, 2007). The composition of the bacterial EPS varies with the type of microorganism (Vaningelgem et al., 2004; Panhota et al., 2007), nutrient availability (De Vuyst & Degeest, 1999; Ricciardi et al., 2002), growth phase and environmental conditions (Fischer et al., 2003; Bahat-Samet et al., 2004). However, the mechanisms involved in the synthesis of EPS seem to be relatively conserved for Gram-negative and Gram-positive bacteria (De Vuyst et al., 2001; Jolly & Stingele, 2001; Laws et al., 2001; Sutherland, 2001; Welman & Maddox, 2003; Whitfield, 2006). In terms of biotechnological applications, bacterial EPS are a good alternative to the polysaccharides of plant and algal origin due to the higher growth rates of the producing bacteria, the reproducible physicochemical properties of their EPS, the easier manipulation of these properties by genetically engineering the producing microorganisms, the presence of novel functionalities and the overall economical costs of producing them (Selbmann et al., 2002; Parikh & Madamwar, 2006). This versatility could be useful in a wide range of applications in many industrial sectors such as textiles, detergents, adhesives, oil recovery, wastewater treatment, dredging, brewing, downstream processing, cosmetology, pharmacology and food additives.

Cyanobacteria are a large and widespread group of photoautotrophic microorganisms that combine the ability to perform oxygenic photosynthesis (similar to that of the chloroplasts) with typical prokaryotic features (Whitton & Potts, 2000). Certain strains also have the ability to fix atmospheric dinitrogen, thus displaying the simplest nutritional requirements (Fay, 1992; Bergman et al., 1997). They possess a unique cell wall that combines the presence of an outer membrane and lipopolysaccharides, as in Gram-negative bacteria, with a thick and highly cross-linked peptidoglycan layer similar to Gram-positive bacteria (Hoiczyck & Hansel, 2000; Stewart et al., 2006). Moreover, many cyanobacterial strains have polysaccharidic structures surrounding their cells (De Philippis & Vincenzini, 2003). However, a lack of information regarding both the genes encoding the proteins involved in the EPS biosynthetic pathways, and the factors controlling these processes strongly limits their potential for biotechnological applications (Morvan et al., 1997; Otero & Vincenzini, 2003; Potts, 2004).

The aim of this review is to summarize the current knowledge on the composition and the macromolecular characteristics of the cyanobacterial exopolysaccharides. We consider their ecological role and the factors that may affect their synthesis, as well as analyze the available genome sequences to gather information on the genes encoding products putatively involved in the production of EPS in cyanobacteria.

Composition and macromolecular characteristics of cyanobacterial exopolysaccharides

The cyanobacterial EPS can be divided in two main groups: the ones associated with the cell surface and the polysaccharides released into the surrounding environment (released polysaccharides, RPS). The EPS associated with the cell surface can be referred to as sheaths, capsules and slimes, according to their thickness, consistency and appearance (De Philippis & Vincenzini, 1998, 2003). The sheath is defined as a thin, dense layer loosely surrounding cells or cell groups usually visible in light microscopy without staining. The capsule generally consists of a thick and slimy layer intimately associated with the cell surface with sharp outlines, which is structurally coherent, excluding particles (e.g. India ink). The slime refers to the mucilaginous material dispersed around the organism but not reflecting the shape of the cells. The RPS are soluble aliquots of polysaccharidic material released into the medium, either from the external layer(s) or derived from a biosynthetic process not directly related to the synthesis of EPS. Despite some evidence, this last point is still controversial. Differences in the sulphur content and in the monosaccharidic composition reported for the sheath and RPS of some cyanobacteria strongly support the hypothesis of different biosynthetic pathways (Tease et al., 1991; Ortega-Calvo & Stal, 1994; Li et al., 2002; Micheletti et al., 2008b). However, further studies are needed to fully elucidate these pathways in cyanobacteria.

The RPS can be easily recovered from liquid cultures and, due to their physicochemical properties, are suitable for a variety of industrial applications, making cyanobacteria one of the most attractive sources of new polymers (De Philippis & Vincenzini, 1998, 2003). The available data on the monosaccharidic composition of cyanobacterial EPS (Table 1) reveal some peculiar features of these polymers when compared with those produced by other microorganisms, such as the presence of one or two uronic acids, constituents rarely found in the EPS produced by other microbial groups. Cyanobacterial EPS also contain sulphate groups, a feature unique among bacteria, but shared by the EPS produced by archaea and eukaryotes. Both the sulphate groups and the uronic acids contribute to the anionic nature of the EPS, conferring a negative charge and a ‘sticky’ behaviour to the overall macromolecule (Decho, 1990; Sutherland, 1994; Leppard et al., 1996; Arias et al., 2003; De Philippis & Vincenzini, 2003; Mancuso Nichols et al., 2005). The anionic charge is an important characteristic for the affinity of these EPS towards cations, notably metal ions. However, the ability to chelate metal ions is related not only to the amount of charged groups but also to their distribution on the macromolecules and their accessibility (Brown & Lester, 1982; De Philippis et al., 2000; Mancuso Nichols et al., 2005; Micheletti et al., 2008b). On the other hand, many cyanobacterial EPS are also characterized by a significant level of hydrophobicity, which is due to the presence of ester-linked acetyl groups (up to 12% of EPS dry weight), peptidic moieties and deoxysugars such as fucose and rhamnose (Table 1). The presence of these hydrophobic groups contributes significantly to the emulsifying properties of the polysaccharides, which would otherwise be highly hydrophilic, and it is also essential for determining their rheological properties (Neu et al., 1992; Shepherd et al., 1995).

Table 1a.   Main constituents of EPS produced by unicellular cyanobacteria
Unicellular organismsEcological originNo. of monosaccharidesWt/EPS dry weight (%)References
  • The cyanobacteria are sourced from a culture collection unless specified otherwise.

  • *

    Field sample.

  • Monocyanobacterial culture.

  • p, present, but not quantified; tr, trace.

Aphanocapsa halophytia MN-11Saline lake, Japan6052.411.9  10.3Sudo et al. (1995)
Aphanothece halophytica GR02Saltworks, China714.127.80  0Li et al. (2001)
Aphanothece sacrum (Sur.) Okada*Freshwater spring, Japan4 p    Kabata et al. (2005)
Chroococcus minutus B 41.79Water reservoir, India144.217.1   3.2Fischer et al. (1997)
Chroococcus submarinusBrackish ponds, French Polynesia101024.414  14Richert et al. (2005)
Cyanothece 16Som2Saltworks, Somalia720.68.2p00 De Philippis & Vincenzini (1998)
Cyanothece CA3Hypersaline ponds, Italy766.823.2p2.70.6 De Philippis & Vincenzini (1998)
Cyanothece CE4Saltworks, Italy780.136.6p0.40.7 De Philippis & Vincenzini (1998)
Cyanothece CE9Saltworks, Italy635.733.5p1.20.0 De Philippis & Vincenzini (1998)
Cyanothece CH1Saltworks, Greece627.439.4p10.5 De Philippis & Vincenzini (1998)
Cyanothece ET2Alkaline lake, Ethiopia863.121.502.34.2 De Philippis & Vincenzini (1998)
Cyanothece ET5Alkaline lake, Ethiopia829.422.300.42.5 De Philippis & Vincenzini (1998)
Cyanothece IR20Hypersaline lake, Israel79.880.2p2.10.8 De Philippis & Vincenzini (1998)
Cyanothece PE13Tidal pool, Greece820.98.9p2.10 De Philippis & Vincenzini (1998)
Cyanothece PE14Tidal pool, Greece721.74.4p0.20.3 De Philippis & Vincenzini (1998)
Cyanothece sp. ATCC 51142Intertidal area, Texas300p  18.3Parikh & Madamwar (2006)
Cyanothece sp. 113Sea, China100   0Chi et al. (2007)
Cyanothece sp. PCC 8801Rice fields, Taiwan83533.3 0.51.4 De Philippis & Vincenzini (1998)
Cyanothece TI4Hypersaline pond, Italy758.239.9p1.41 De Philippis & Vincenzini (1998)
Cyanothece TP10Saltworks, Italy731.342.7p3.9  De Philippis & Vincenzini (1998)
Cyanothece TP5Saltworks, Italy640.457.5p1.10 De Philippis & Vincenzini (1998)
Cyanothece VI13Tidal pool, Italy832.137.2p0.30 De Philippis & Vincenzini (1998)
Cyanothece VI22Tidal pool, Italy740.831.7p0.20.6 De Philippis & Vincenzini (1998)
Gloeocapsa gelatinosaUnknown1031.414.2   28.2Raungsomboon et al. (2006)
Gloeothece sp. PCC 6909Unknown65313.8  6.2Tease et al. (1991)
Johannesbaptistia pellucidaBrackish ponds, French Polynesia957.619  10Richert et al. (2005)
Microcystis aeruginosaWater reservoir, Brazil88.928.3    Gouvea et al. (2005)
Microcystis aeruginosa f. aeruginosaUnknown45.2 μg mg−164.6    Forni et al. (1997)
Microcystis aeruginosa f. flos-aquaeUnknown32.6 μg mg−10    Forni et al. (1997)
Microcystis flos-aquae C3-40Unknown6835.5   <1Plude et al. (1991)
Microcystis PCC 7005Lake, Wisconsin40.05 μg mg−115.7    Forni et al. (1997)
Microcystis PCC 7941Lake, Canada40.04 μg mg−1tr    Forni et al. (1997)
Microcystis sp.Water reservoir, Taiwan5Pp    Huang et al. (2007)
Microcystis viridisUnknown65.8 μg mg−17.3    Forni et al. (1997)
Microcystis wesenbergiiUnknown11000    Forni et al. (1997)
Rhabdoderma rubrumBrackish ponds, French Polynesia828.713  7Richert et al. (2005)
Synechococcus elongatus f. A. nidulansLake, Ohio300    Sangar & Dugan (1972)
Synechocystis sp. PCC 6714Freshwater, California1116.731.41.20 20Panoff et al. (1988)
Synechocystis sp. PCC 6803Freshwater, California1216.46.71.00 40Panoff et al. (1988)
Table 1b.   Main constituents of EPS produced by filamentous cyanobacteria
Filamentous organismsEcological originNo. of monosaccharidesWt/EPS dry weight (%)References
Uronic acidsDeoxysugarsSulphatePyruvateAcetatePeptides
  • The cyanobacteria are sourced from a culture collection unless specified otherwise.

  • *

    Monocyanobacterial culture.

  • Field sample.

  • p, present, but not quantified; tr, trace.

Geitlerinema sp.*Brackish ponds, French Polynesia46tr0  8Richert et al. (2005)
Leptolyngbya sp. VRUC 135Domus Aurea, Italy503    Piro et al. (2005)
Lyngbya confervoides S9gSurface of Lithophyllum lichenoides, France938.6tr    Gloaguen et al. (1995)
Microcoleus sp.*Sand dunes, Israel50p   6.0Mazor et al. (1996)
Microcoleus vaginatus*Desert algal crusts, China1089.9   50.3Hu et al. (2003a)
Oscillatoria amphibia PCC 7105Unknown marine habitat86.74.59.1   Gloaguen et al. (1995)
Oscillatoria corallinae CJ1Leaves of Posidonia Oceanica, France924.25.119.3   Gloaguen et al. (1995)
Oscillatoria sp.Microbial mats, Florida104.615.9    Bender et al. (1994)
Oscillatoria sp.Pantelleria Island, Italy5p1    Nicolaus et al. (1999)
Oscillatoria sp.Contaminated soil, India3000  34.4Parikh & Madamwar (2006)
Oscillatoria sp.Water reservoir, Taiwan30p    Huang et al. (2007)
Phormidium ectocarpi ME3Marine habitat918.92.715.2   Gloaguen et al. (1995)
Phormidium ectocarpi C86Marine habitat729.9tr12.2   Gloaguen et al. (1995)
Phormidium ectocarpi K5Marine habitat928.71.812.4   Gloaguen et al. (1995)
Phormidium ectocarpi N182Marine habitat628.7tr13.6   Gloaguen et al. (1995)
Phormidium ectocarpi PCC7375Marine plankton, Massachusetts841.510.4    Gloaguen et al. (1995)
Phormidium foveolarumunknown9p17.4    Matulewicz et al. (1984)
Phormidium foveolarum C52Marine habitat929.4211.2   Gloaguen et al. (1995)
Phormidium foveolarum MEUFreshwater80.59.80.5   Gloaguen et al. (1995)
Phormidium minutum D5Marine habitat717.10.94.5   Gloaguen et al. (1995)
Phormidium minutum NB5Marine habitat724.4tr    Gloaguen et al. (1995)
Phormidium minutum RT6Marine habitat920.17.17.7   Gloaguen et al. (1995)
Phormidium sp.Lake, Antarctica9924.5   13Matulewicz et al. (1984)
Phormidium sp.Unknown6 5.8    Nicolaus et al. (1999)
Phormidium sp. 90-14/1Intertidal habitat, New Guinea83.05.83   Gloaguen et al. (1995)
Phormidium sp. CCAP1463/4Phosphorescent bay, Massachusetts913.06.73.5   Gloaguen et al. (1995)
Phormidium sp. CCAP1464/3Marine habitat826.1tr4.4   Gloaguen et al. (1995)
Phormidium sp. J-1Benthic, drainage canal, Israel 34.0 1.65  4.4Bar-Or & Shilo (1987)
Phormidium sp. PNG91Intertidal habitat, New Guinea9tr5.63.6   Gloaguen et al. (1995)
Phormidium tenue*Desert algal crusts, China8tr12.7   21.9Hu et al. (2003a)
Phormidium uncinatumUnknown50<50   Hoiczyk (1998)
Plectonema battersii*Brackish ponds, French Polynesia8711.116  11Richert et al. (2005)
Plectonema golenkinianum*Brackish ponds, French Polynesia81120.40  19Richert et al. (2005)
Spirulina maximaChina8pp    Nie et al. (2002)
Spirulina platensisChina520.0p    Tseng & Zhao (1994)
Spirulina platensis PCC 8005Unknown104013.65   Filali Mouhim et al. (1993)
Table 1c.   Main constituents of EPS produced by filamentous heterocystous cyanobacteria
heterocystous organisms
Ecological originNo. of monosaccharidesWt/EPS dry weight (%)References
Uronic acidsDeoxysugarsSulphatePyruvateAcetatePeptides
  • The cyanobacteria are sourced from a culture collection unless specified otherwise.

  • *

    Monocyanobacterial culture.

  • Field sample.

  • p, present, but not quantified; tr, trace.

Anabaena cylindricaUnknown500    Dunn & Wolk (1970)
Anabaena cylindrica 10 CUnknown71.710.1p*   Lama et al. (1996)
Anabaena cylindrica CCAP1403/2Freshwater pond, UK6286    Bishop et al. (1954)
Anabaena flos-aquaeFreshwater4p0    Moore & Tischer (1964)
Anabaena flos-aquae A-37Unknown458.40    Wang & Tischer (1973)
Anabaena flos-aquae A-37Unknown40.90    Moore & Tischer (1965)
Anabaena sp.Water reservoir, Taiwan30p    Huang et al. (2007)
Anabaena sp. ATCC 33047Algal mat, Texas519.400  7Moreno et al. (2000)
Anabaena sp. C5*Soil, Yugoslavia5p    51.5Gantar et al. (1995)
Anabaena sp. PC-1Unknown 5 p  12Choi et al. (1998)
Anabaena sphaericaUnknown4p0    Nicolaus et al. (1999)
Anabaena spiroides*Water reservoir, Brazil8933.9    Gouvea et al. (2005)
Anabaena torulosaSoil, France708.7    Nicolaus et al. (1999)
Anabaenopsis circularis PCC 6720Drainage canal, Israel 0 0  0Bar-Or & Shilo (1987)
Chlorogloeopsis sp. PCC 6912Soil, India6p10.2    Nicolaus et al. (1999)
Cyanospira capsulataAlkaline lake, Kenya737.512.5 p pGarozzo et al. (1995)
Cyanospira capsulata ATCC 43193Alkaline lake, Kenya536.515.401.502.0Vincenzini et al. (1990)
Fischerella maior Nav 10 bisCatacombs, Italy712.821.4p   Bellezza et al. (2006)
Fischerella muscicolaUnknown5 17.6    Nicolaus et al. (1999)
Mastigocladus laminosusThermal spring, France11pp    Gloaguen et al. (1999)
Nostoc flagelliformeDesert China300    Huang et al. (1998)
Nostoc calcicola 79WA01Soils, Washington921.815.5   30Flaibani et al. (1989)
Nostoc carneumContaminated soil, India2000  27.2Parikh & Madamwar (2006)
Nostoc communeMountain area, China713.35.8    Huang et al. (1998)
Nostoc communeFreshwater, China618.40    Brüll et al. (2000)
Nostoc commune DRH-1Desert, Mongolia722.700   Helm et al. (2000)
Nostoc commune UTEX584Unknown9429    Flaibani et al. (1989)
Nostoc insulare 54.79Soil426.400trtr0.7Volk et al. (2007)
Nostoc insulare 54.79Soil825.38.9   3.5Fischer et al. (1997)
Nostoc linckia f. muscorumUnknown6pp    Mehta & Vaidya (1978)
Nostoc sp.Contaminated soil, India2000  40.1Parikh & Madamwar (2006)
Nostoc sp.Unknown4pp    Moore & Tischer (1964)
Nostoc sp.Unknown6pp    Mehta & Vaidya (1978)
Nostoc sp.*Desert algal crusts, China603.5   7.5Hu et al. (2003a)
Nostoc sp. 221Unknown341.30    Mehta & Vaidya (1978)
Nostoc sp. 2S9B*Soil, Yugoslavia4p    2.8Gantar et al. (1995)
Nostoc sp. DUnknown600    Cupac & Gantar (1992)
Nostoc sp. PCC 6302Unknown869.43.9000.56.9De Philippis et al. (2000)
Nostoc sp. PCC 6310Pond, Israel94.31.6p1.43.369.5De Philippis et al. (2000)
Nostoc sp. PCC 6705Botanical garden, California926.82.1p3.21.76.9De Philippis et al. (2000)
Nostoc sp. PCC 6719Soil water culture, California99.90.7p5.02.40.8De Philippis et al. (2000)
Nostoc sp. PCC 6720Soil, Indonesia813.54.1p2.11.13.1De Philippis et al. (2000)
Nostoc sp. PCC 7107Pond, California926.76.7p0.72.73.4De Philippis et al. (2000)
Nostoc sp. PCC 7119Unknown65.20.5p5.03.40.9De Philippis et al. (2000)
Nostoc sp. PCC 7413Soil, UK919.32.0p6.23.30.6De Philippis et al. (2000)
Nostoc sp. PCC 7416Shallow pool, California823.64.6p04.09.7De Philippis et al. (2000)
Nostoc sp. PCC 7422Cycas roots826.85.4p2.42.04.7De Philippis et al. (2000)
Nostoc sp. PCC 7423Dried soil, Senegal718.80.8p3.26.818.4De Philippis et al. (2000)
Nostoc sp. PCC 7706Water below calcareous stone, France725.21.2000.810.3De Philippis et al. (2000)
Nostoc sp. PCC 7803Sand dunes, UK822.00.800.21.15.5De Philippis et al. (2000)
Nostoc sp. PCC 7807Soil, France728.70.6p6.12.010.2De Philippis et al. (2000)
Nostoc sp. PCC 7906Freshwater,837.73.3001.915.6De Philippis et al. (2000)
Nostoc sp. PCC 7933Mud, Finland833.74.8p01.45.9De Philippis et al. (2000)
Nostoc sp. PCC 7936Rice field, India551.10.6p5.81.82.1De Philippis et al. (2000)
Nostoc sp. PCC 7937Freshwater, Mississippi74.90.9p3.53.03.2De Philippis et al. (2000)
Nostoc sp. PCC 8009/1Coralloid roots of Macrozamia lucida812.71.2p0.43.73.3De Philippis et al. (2000)
Nostoc sp. PCC 8109Unknown762. Philippis et al. (2000)
Nostoc sp. PCC 8112Laundromat discharge pond, Michigan942.25.0p0.40.911.7De Philippis et al. (2000)
Nostoc sp. PCC 8113Unknown86.65.1p5.803.8De Philippis et al. (2000)
Nostoc sp. PCC 8306Soil, West Africa756.73.200.7021.7De Philippis et al. (2000)
Nostoc sp. PCC 9202Rice field, Spain937.72.00005.0De Philippis et al. (2000)
Nostoc sp. PCC 9305Anthoceros, California813.24.402.304.5De Philippis et al. (2000)
Nostoc sp. WV2Unknown6pp    De Philippis & Vincenzini (1998)
Scytonema hofmanni PCC 7110Cave (limestone), Bermuda3 0    Nicolaus et al. (1999)
Scytonema javanicum*Desert algal crusts, China10tr7.4   50.2Hu et al. (2003a)
Scytonema ocellatum CP8-2Catacombs, Italy716.612.5p   Bellezza et al. (2006)
Scytonema sp.Thallus of D. glabratum, Brazil10 11    Sassaki et al. (2005)
Tolypothrix tenuis PCC 7101Soil, Borneo6 14.3    Nicolaus et al. (1999)

Cyanobacterial EPS are complex heteropolysaccharides, with c. 75% of the polymers described so far composed of six or more different kinds of monosaccharides. This feature contrasts with the polymers synthesized by other bacteria or macroalgae, which contain a lower number of different monomers, usually less than four (De Philippis & Vincenzini, 1998). To date, up to 12 different monosaccharides have been identified in cyanobacterial EPS (Table 1): the hexoses, glucose, galactose, mannose and fructose, the pentoses, ribose, xylose and arabinose, the deoxyhexoses, fucose, rhamnose and methyl rhamnose, and the acidic hexoses, glucuronic and galacturonic acid (De Philippis & Vincenzini, 1998, 2003; De Philippis et al., 2001). In a few cases, the presence of additional types of monosaccharides (i.e. methyl sugars and/or amino sugars) such as N-acetyl glucosamine, 2,3-O-methyl rhamnose, 3-O-methyl rhamnose, 4-O-methyl rhamnose and 3-O-methyl glucose have been reported (Hu et al., 2003a). The monosaccharide most frequently found at the highest concentration in cyanobacterial EPS is glucose, although there are polymers where other sugars, such as xylose, arabinose, galactose or fucose, are present at higher concentrations than glucose (Tease et al., 1991; Bender et al., 1994; Gloaguen et al., 1995; Fischer et al., 1997; De Philippis & Vincenzini, 1998, 2003; Parikh & Madamwar, 2006).

The high number of different monosaccharides found in cyanobacterial EPS and the consequential variety of linkage types is usually considered a reason for the presence of complex repeating units, as well as for a broad range of possible structures and architectures of these macromolecules. As one consequence of this complexity, the cyanobacterial EPS are less well characterized than those of other microorganisms and only a few structures have been proposed (Table 2). The polysaccharides produced by Nostoc commune DRH-1, Nostoc insulare and Cyanothece sp. ATCC 51142 are composed of repeating units of six, four and three monosaccharides, respectively (Helm et al., 2000; Huang et al., 2000; Shah et al., 2000; Volk et al., 2007). On the other hand, the structures proposed for the EPS produced by Mastigocladus laminosus and Cyanospira capsulata are far more complex, with repeating units of 15 and eight monosaccharides, respectively (Garozzo et al., 1995, 1998; Gloaguen et al., 1995, 1999). For Spirulina platensis, no structure was proposed, but it was demonstrated that its EPS repeating unit contains at least 15 sugar residues (Filali Mouhim et al., 1993).

Table 2.   Overview of the published structures of the heteropolysaccharides produced by cyanobacteria
Organisms (ecological origin)Repeating unitsReferences
  • The cyanobacteria are sourced from a culture collection unless specified otherwise.

  • GlcA, glucuronic acid; GalA, galacturonic acid; NosA, nosturonic acid; Ido, idose; Glc, glucose; Gal, galactose; Fuc, fucose; Ara, arabinose; Man, mannose; Rha, rhamnose; Xyl, xylose; Rib, ribose. p, pyranose form; f, furanose form; α, anomer α; β, anomer β.

  • *

    Field sample.

 Cyanothece sp. ATCC 51142 (intertidal area, Texas)inline imageShah et al. (2000)
Filamentous heterocystous
 Cyanospira capsulate (unknown)inline imageGarozzo et al. (1995, 1998)
 Mastigocladus laminosus* (thermal spring, France)inline imageGloaguen et al. (1995)
 Nostoc commune DRH-1 (desert, Mongolia)inline imageHelm et al. (2000), Huang et al. (2000)
 Nostoc insulare 54.79 (soil)inline imageVolk et al. (2007)

The knowledge of the structure of a polysaccharide is generally considered necessary to infer its physicochemical properties (De Philippis & Vincenzini, 1998). Indeed, the interest in cyanobacteria as producers of high-molecular-weight polysaccharides is related to the capability of these biopolymers to modify the rheological properties of water, acting as thickening agents (Sutherland, 1996), and to stabilize the flow properties of aqueous solutions. Thus, one of the key features of a polysaccharide, which determines most of the properties generally considered to be useful for industrial applications, is its high molecular mass (Shepherd et al., 1995), as this characteristic has a direct influence on the rheological properties of solutions of the polymer (Kamal et al., 2003). The molecular masses reported thus far for the exopolysaccharides released by cyanobacteria are listed in Table 3; the highest molecular masses were found for the polysaccharides produced by C. capsulata, Anabaena spiroides and Phormidium 94, which are about 2 MDa. These values, significantly higher than that of xanthan gum, which has a molecular mass of about 1 MDa (Kamal et al., 2003), point to the potential of these polymers for biotechnological exploitation as viscosifying or suspending agents. In this context, it is worth mentioning that the viscosity of some of the cyanobacterial exopolysaccharides is comparable to, or even higher than, that of aqueous solutions of xanthan gum at similar concentrations (Sutherland, 1996; De Philippis et al., 2000). However, even if only a very limited number of cyanobacterial EPS have been fully described regarding their flow properties (Cesàro et al., 1990; Navarini et al., 1990; Lapasin et al., 1992; Moreno et al., 2000; Morris et al., 2001), there are a few reports emphasizing the dependence of viscosity on the shear rate of water solutions of cyanobacterial EPS and of commercial xanthan gum. Comparing the viscosity data (Table 4), it is possible to conclude that some of the RPS produced by cyanobacteria (e.g. the RPS synthesized by Cyanothece strains CE4 and CA3) possess very high viscosities, up to four times higher than that of xanthan gum. However, it should be stressed that, to make a reliable comparison of the flow properties of two polysaccharides, it is necessary to evaluate additional rheological properties, as well as to assess the dependence of these properties on factors such as pH, temperature and the ionic strength of the solution.

Table 3.   Molecular masses of the EPS released by cyanobacteria
OrganismEcological originApparent molecular mass (kDa)References
  • The cyanobacteria are sourced from a culture collection unless specified otherwise.

  • *

    Monocyanobacterial culture.

  • Field sample.

 Chroococcus minutus B 41.79Water reservoir, India1200–1600Fischer et al. (1997)
 Microcoleus vaginatus*Desert algal crusts, China380Hu et al. (2003a)
 Oscillatoria sp.Microbial mats, Florida≥200Bender et al. (1994)
 Phormidum 94a*Arid soil, Mexico2000Vicente-Garcia et al. (2004)
 Phormidium J-1Drainage canal, Israel1200Bar-Or & Shilo (1987)
 Phormidum tenue*Desert algal crusts, China380Hu et al. (2003a)
 Spirulina platensisChina81–98Tseng & Zhao (1994)
Filamentous heterocystous
 Anabaena circularis PCC 6720Drainage canal, Israel>1200Bar-Or & Shilo (1987)
 Anabaena spiroides*Water reservoir, Brazil2000Colombo et al. (2004)
 Anabaena sp. ATCC 33047Algal mat, Texas1350Moreno et al. (2000)
 Cyanospira capsulata ATCC 43193Alkaline lake, Kenya1400–1900Vincenzini et al. (1993), Cesàro et al. (1990)
 Nostoc insulare 54.79Soil540–1300Fischer et al. (1997)
 Nostoc sp.*Desert algal crusts, China460Hu et al. (2003a)
 Schizothrix sp.*Intertidal marine stromatolites, Bahamas300Kawaguchi & Decho (2002)
 Scytonema javanicum*Desert algal crusts, China110–380Hu et al. (2003a)
Table 4.   Viscosity (expressed as mPa s and measured at 10.1 s−1 shear rate) of 0.1% (w/v) water solutions of pure cyanobacterial polysaccharides (RPS) or of xanthan gum (Kelco Keltrol, commercial grade)
(mPa s)
  • *

    Exopolysaccharides in a 0.4% (w/w) solution.

  • Exopolysaccharides in a 0.6% (w/v) solution.

  • Intrinsic viscosity (μ) in deionized water.

Anabaena ATCC 33047100.0*Moreno et al. (2000)
Aphanothece halophytia GR029.5Morris et al. (2001)
Cyanospira capsulata158.5De Philippis & Vincenzini (1998)
Cyanothece CA3398.1De Philippis & Vincenzini (1998)
Cyanothece CE4400.0De Philippis et al. (2001)
Cyanothece ET25.6De Philippis & Vincenzini (1998)
Cyanothece IR2080.0De Philippis & Vincenzini (1998)
Cyanothece PE14158.5De Philippis & Vincenzini (1998)
Nostoc cameum6.9Parikh & Madamwar (2006)
Nostoc PCC 670540.0De Philippis et al. (2001)
Nostoc PCC 670519.5De Philippis et al. (2000)
Nostoc PCC 7119125.9De Philippis et al. (2000)
Nostoc PCC 74227.9De Philippis et al. (2000)
Nostoc PCC 7423158.5De Philippis et al. (2000)
Nostoc PCC 793712.3De Philippis et al. (2000)
Nostoc sp.11.7Parikh & Madamwar (2006)
Oscillatoria sp.12.1De Philippis et al. (2000)
Xanthan gum78.9De Philippis & Vincenzini (1998)

It has been reported that cyanobacterial EPS are not only composed of carbohydrates but also of other macromolecules such as polypeptides (Kawaguchi & Decho, 2000). Polypeptides enriched with glycine, alanine, valine, leucine, isoleucine and phenylalanine have been reported in the EPS of C. capsulata and Nucula calcicola (Flaibani et al., 1989; Marra et al., 1990), and in Schizothrix sp., small proteins specifically enriched with aspartic and glutamic acid have been observed (Kawaguchi & Decho, 2002). In general, the chemical composition, the type and the amount of the exopolysaccharides produced by a given cyanobacterial strain are stable features, mostly depending on the species and the cultivation conditions (Nicolaus et al., 1999). However, the sugar composition of the EPS produced by a certain strain may, qualitatively and quantitatively, vary slightly, especially with the age of the culture (Gloaguen et al., 1995; De Philippis & Vincenzini, 1998).

According to the chemical and physicochemical features of the cyanobacterial exopolysaccharides summarized above, there are at least three possible fields of application for these polymers: (1) in the food, cosmetic, textile or painting industries, for the modification of the flow properties of water, i.e. as thickening, suspending or emulsifying agents (De Philippis & Vincenzini, 1998, 2003; Li et al., 2002; Parikh & Madamwar, 2006); (2) in the pharmaceutical industry, because of their antiviral or immuno-stimulating properties or the capability of slowly releasing drugs (Schaeffer & Krylov, 2000; Jensen et al., 2001; Pugh et al., 2001; Ghosh et al., 2009); (3) in waste water treatment plants or the goldsmith industry, for the chelation of toxic or valuable metal ions from water solutions, i.e. as biosorbents (De Philippis & Micheletti, 2009).

Considering the extensive literature claiming the potential for industrial exploitation of these biopolymers, one would expect that at least for some of them the technology transfer had already occurred. However, in spite of a significant number of patents available, covering the use of cyanobacterial polysaccharides in various industrial fields (see, for instance, the review published in 2006 by Sekar & Paulraj on the patents filed at the US Patent and Trademark Office), no industrial product derived from these biopolymers is available in the market. In our opinion, the main reason for this discrepancy is the presence in the market of well-established industrial processes for heterotrophic microorganisms that in the short term would be expensive to convert for cyanobacteria. In the case of thickening agents in foods, it has to be stressed that there are already other microbial polysaccharides in the market, the most important being xanthan gum, gellan and pullulan, respectively, produced by Xanthomonas campestris, Pseudomonas elodea and Aureobasidium pullulans, and dextran, produced by several lactic acid bacteria belonging to the genera Leuconostoc, Lactobacillus and Streptococcus. These biopolymers have already undergone the complex, expensive and time-consuming procedures for their approval as food additives. Thus, even if some of the cyanobacterial exopolysaccharides, such as the one produced by C. capsulata (Navarini et al., 1990), show better rheological properties, the differences would not be significant enough to risk a new technology transfer in competition with well-established commercial products. Similarly, the exploitation of cyanobacteria producing polysaccharides with good antiviral activity has not been considered worth developing new drugs. This is due to the long and very expensive procedures needed for the commercialization of new pharmaceutical products.

The possible use of exopolysaccharide-producing cyanobacteria for the recovery of valuable metals from industrial wash waters seems to be more promising than most of the above-mentioned applications. Indeed, the high economical value of the metal, which can be easily recovered from the biosorbent, might justify the investment necessary for the production of the biomass. However, this field of application is still in its infancy and needs more research to establish a simple and cheap technology for the production and utilization of the cyanobacterial biomass as biosorbent, as well as for the recovery of the metal.

Putative roles of EPS in cyanobacteria

The capability of cyanobacteria to survive in severe habitats (e.g. at the surface of a sand dune in a desert or exposed to high UV radiation on the lithic surfaces of monuments) has been related to the protective mechanisms that they have developed. Among these mechanisms, one of the most diffused within the cyanobacteria is the ability to synthesize external polysaccharidic layers that protect the cells from unfavourable environmental conditions. Many studies, in fact, have shown that a coating of extracellular polysaccharidic material can protect bacteria against dehydration, phagocytosis, antibody recognition and even lysis by viruses (Dudman, 1977; Tease & Walker, 1987; Hill et al., 1994; Scott et al., 1996; Hoiczyk, 1998; De Vuyst & Degeest, 1999; Sutherland, 1999; Ruas-Madiedo et al., 2002; De Philippis & Vincenzini, 2003; Tsuneda et al., 2003; Welman & Maddox, 2003), or confer to the cells the ability to adhere to a solid substrate, preventing them from being washed away from their natural habitat by water flow (Dudman, 1977; Scott et al., 1996; De Philippis et al., 2005).

A number of cyanobacteria are capable of surviving nearly without water, producing both internal and external polysaccharides, which help to stabilize the macromolecular constituents of the cell, as well as the cell structure. It has been suggested that these polysaccharides can form hydrogen bonds with proteins, lipids and DNA, thus replacing the water shell that usually surrounds these macromolecules (Potts, 1994). EPS, owing to their hydrophilic/hydrophobic characteristics (see previous section), are able to trap and accumulate water, creating a gelatinous layer around the cells that regulates water uptake and loss, and stabilizes the cell membrane during periods of desiccation (Grilli Caiola et al., 1993, 1996; Tamaru et al., 2005). Upon rehydration, cyanobacteria can rapidly recover metabolic activities and repair cellular components (Scherer et al., 1984, 1986; Satoh et al., 2002; Fleming & Castenholz, 2007). A good example of this is the filamentous EPS-producing cyanobacterium N. commune, which is ubiquitously distributed from the tropics to the polar regions of the Earth. These cyanobacteria form macroscopic colonies in which the entangled filaments are embedded in massive polysaccharidic structures. In their natural environment, these colonies are subjected to frequent desiccation and rewetting cycles, during which they release large quantities of protective proteins and compounds such as mycosporine-like amino acids, UV-screen pigments and active Fe-containing superoxide dismutase (Hill et al., 1994; Böhm et al., 1995; Shirkey et al., 2000).

Another important consequence of the above-described hydrophobicity of the cyanobacterial EPS becomes evident in desert microbial crusts, where the polysaccharides contribute to the hydrological properties of the soil by clogging sand particles and by causing the run-off of water on the dune, protecting the microbial community of the crusts from being washed away by the water flow (Mazor et al., 1996; Kidron et al., 1999).

Recently, it was demonstrated that the cyanobacterial sheath can protect the cells from the detrimental process of biomineralization (Phoenix et al., 2000; Benning & Mountain, 2004). In fact, permeability studies demonstrated that the sheath of Calothrix sp. was impermeable to particles of at least 11 nm diameter, thus preventing the colloids from biomineralizing the sensitive components of the cell wall (Phoenix et al., 2000; Benning et al., 2004).

Furthermore, the presence of negatively charged polysaccharidic layers surrounding cyanobacterial cells may play an important role in the sequestration of metal cations, and in creating a microenvironment enriched in those metals that are essential for cell growth but are present at very low concentrations in some environments (Parker et al., 1996; Sutherland, 1999). On the other hand, the presence of a polysaccharidic layer surrounding the cells can also prevent direct contact between the cells and toxic heavy metals that may be present in the environment. Actually, it was recently suggested that the high viscosity of the cultures of C. capsulata, due to the solubilization in the culture medium of large amounts of a high molecular mass RPS, hindered the free diffusion of copper ions into the culture (De Philippis et al., 2007).

The UV-absorbing pigment scytonemin was found in the sheath of a number of cyanobacteria living in environments characterized by a high level of solar irradiation (Garcia-Pichel & Castenholz, 1991; Ehling-Schulz et al., 1997; Ehling-Schulz & Scherer, 1999). Moreover, in the sheath of some cyanobacterial strains, mycosporine-like amino acid compounds (MMAs) were also found (Adhikary & Sahu, 1998), confirming the role of the sheath in harbouring UV-absorbing substances, and thus protecting the cyanobacterial cells from the deleterious effects of UV radiation.

EPS may also play an important role in the locomotion of gliding cyanobacteria. Indeed, the secretion of slime can provide the necessary propulsive force for movement (Li et al., 2002). Cyanobacterial exopolysaccharides may also protect nitrogenase (the complex responsible for nitrogen fixation) from the deleterious effects of oxygen (Kallas et al., 1983).

Only a few of the ecological roles attributed to the cyanobacterial exopolysaccharides are fully supported by experimental data or detailed ecological observations. For instance, their role in protecting the cells against desiccation was experimentally demonstrated by a number of authors, in particular by Malcom Potts' group (Potts, 1994, 1999, 2004; Shaw et al., 2003; Wright et al., 2005) and Tamaru et al. (2005). These publications described in detail some of the chemical and molecular mechanisms by which the exocellular polysaccharidic layers prevent possible cell damage caused by desiccation and rewetting processes. Moreover, there is evidence that most of the cyanobacteria isolated from very dry environments (desert soils, lithic surfaces of monuments located in arid environments, etc.) are characterized by the capacity to synthesize large amounts of exocellular polysaccharidic material (Danin et al., 1998; Brüll et al., 2000; Belnap & Lange, 2001; Hu et al., 2003b; Crispim & Gaylarde, 2005; Rivera-Aguilar et al., 2006; Zhang et al., 2006), thus supporting the role of these macromolecules in the survival of cyanobacteria in arid habitats. Another possible role that has been thoroughly investigated is the capacity of the sheath, capsules and slime to protect the cyanobacterial cells from the harmful effects of UV radiations. It was demonstrated that UV irradiation induces the synthesis of the extracellular polysaccharidic matrix in N. commune (Ehling-Schulz et al., 1997; Wright et al., 2005) and also that the UV-screen pigments are accumulated in the sheath and in the extracellular matrix, constituting a barrier against the penetration of the harmful UV radiations (Böhm et al., 1995; Ehling-Schulz & Scherer, 1999; Dillon et al., 2002; Fleming & Castenholz, 2007, 2008).

On the other hand, the observed capacity of the cyanobacterial exopolysaccharides to chelate metal ions has been reported to enable cells to accumulate the metals necessary for their growth and/or to prevent cells from direct contact with metals with toxic effects. Indeed, this assumption arises from experiments demonstrating that most cyanobacterial EPS are anionic in nature due to the presence of charged constituents, such as uronic acids, sulphate and ketal-linked pyruvate groups (Table 1). Additionally, many studies demonstrated the affinity of exopolysaccharides for metals (Micheletti et al., 2008a; De Philippis & Micheletti, 2009). However, direct experimental evidence demonstrating the ecological role of this metal-uptake capacity is not yet available.

The role of the cyanobacterial polysaccharidic investments seems to differ from strain to strain, and to be dependent on the physical and chemical characteristics of the natural habitat or culture medium in which the organism grows. A more accurate perception of the ecological roles of these polymers will be possible when the information on the genetic machinery related to their production is available. This will enable the conditions under which the genes are transcribed/expressed to be investigated.

Factors affecting biosynthesis of cyanobacterial EPS

The use of cyanobacterial EPS for biotechnological applications depends on the identification of culture parameters that influence the synthesis and/or the characteristics of the EPS, and, subsequently, the establishment and control of the conditions that optimize the productivity and the suitable characteristics of the polymer. During the last three decades, several main factors controlling the production of the cyanobacterial EPS have been identified. These include energy availability and the C : N ratio (De Philippis & Vincenzini, 1998; Li et al., 2002). However, other important factors such as the amounts of other nutrients as well as growth parameters such as light intensity, salinity and temperature have been largely disregarded, and very few exhaustive studies on factors influencing the production of cyanobacterial EPS are available in the literature. Moreover, the responses of cyanobacteria to changes in culture conditions appear to be frequently strain-dependent, making the optimization of EPS production even more difficult. The known key factors affecting EPS production are summarized in Table 5.

Table 5.   Effects of culture conditions on the EPS production in cyanobacteria
of combined
in NaCl
air flow
Increase in
in light
  1. +, positive effect (increased production); , negative effect (decreased production); =, no changes.

 Aphanocapsa halophyta MN11++   +Sudo et al. (1995), Matsunaga et al. (1996)
 Cyanothece sp. 113+ ++ ++Su et al. (2007)
 Cyanothece sp. 16Som2+=    De Philippis et al. (1993), De Philippis & Vincenzini (1998)
 Cyanothece sp. ATCC 51142+ +    Nicolaus et al. (1999), Shah et al. (1999)
 Gloeocapsa gelatinosa+    ++Raungsomboon et al. (2006)
 Gloeothece sp. ATCC 27152+      Tease & Walker (1987)
 Synechococcus elongatus f. A. nidulans+      Sangar & Dugan (1972)
 Synechococcus sp.++=    Roux (1996)
 Synechococcus sp. BG0011     + De Philippis & Vincenzini (1998)
 Microcoleus vaginatus+ +    Hu et al. (2003a), Chen et al. (2006)
 Phormidium laminosum (OH-1-pCl1)      Fresnedo & Serra (1992)
 Phormidium sp.+ + + Nicolaus et al. (1999)
 Phormidium tenue+      Hu et al. (2003a)
 Spirulina sp.++   Nicolaus et al. (1999)
Filamentous heterocystous
 Anabaena cylindrica 10 C+     Lama et al. (1996)
 Anabaena flos-aquae A37=      Tischer & Davis (1971)
 Anabaena sp. ATCC 33047 ++ +Moreno et al. (1998)
 Anabaena sp. PC-1+ +    Choi et al. (1998)
 Anabaena sp. WSAF+ + + Nicolaus et al. (1999)
 Anabaena sp.++     Huang et al. (2007)
 Anabaena torulosa+ + + Nicolaus et al. (1999)
 Cyanospira capsulata ==  + De Philippis et al. (1991)
 Nostoc commune+      Huang et al. (1998)
 Nostoc sp.      Hu et al. (2003a)
 Nostoc sp. PCC 7413+     +Otero & Vincenzini (2003)
 Nostoc sp. PCC 7936   = +Otero & Vincenzini (2003, 2004)
 Nostoc sp. PCC 8113     +Otero & Vincenzini (2003, 2004)
 Scytonema javanicum      Hu et al. (2003a)
 Westiellopsis prolifica ARM 366  +    Jha et al. (1987)


Nitrogen is one of the most important elements for the synthesis of cell material, and cyanobacteria are either dependent on a combined nitrogen source or, in a restricted number of strains, can fix atmospheric nitrogen. Correlation between the source/amount of nitrogen and the production of EPS has been evaluated for several cyanobacteria and different results were observed depending on the strain tested. Usually, as can be observed in Table 5, the presence of a combined nitrogen source in the culture medium resulted in an increase in EPS synthesis, probably due to the lower energy requirement necessary for the assimilation of combined nitrogen compared with the energy needed for nitrogen fixation (Otero & Vincenzini, 2003; Kumar et al., 2007). In some cyanobacteria, the amount of polymer produced varied according to the nitrogen source used (De Philippis & Vincenzini, 1998), whereas Anabaena flos-aquae A37 showed similar EPS production when supplied with different nitrogen sources such as Mg(NO3)2, KNO3, NaNO3, NH4NO3 and NH4Cl (Tischer & Davis, 1971). Moreover, it was also demonstrated that the composition of the polymer released by Anabaena cylindrica 10C was slightly modified when the strain was cultivated with different nitrogen sources (De Philippis & Vincenzini, 1998). Nitrogen starvation has often been described as a condition that enhances EPS synthesis (De Philippis et al., 1993; Otero & Vincenzini, 2003), probably because this contributes to the increase in the C : N ratio, thus promoting the incorporation of carbon into polymers (Otero & Vincenzini, 2003; Kumar et al., 2007). Nevertheless, it is difficult to detect a direct correlation between diazotrophic and nitrogen-limiting conditions because other factors, such as differences in the carbon fixation efficiency and in the control of the equilibrium between internal and extracellular carbon pools, may explain the variations observed in the production of EPS under different culture conditions (De Philippis & Vincenzini, 1998; Otero & Vincenzini, 2003). Indeed, it was observed that in the nitrogen-fixing cyanobacterium C. capsulata, the mere diversion of carbon flux from protein synthesis, caused by the addition of various inhibitors of nitrogen assimilation, induced the accumulation of intracellular carbohydrate reserves (i.e. glycogen), whereas an effective enhancement of the amount of carbon available to the cells, induced by the addition of glyoxylate, which is known to stimulate the CO2 fixation rate, caused an increase in the amount of EPS synthesized and released by the cells (De Philippis et al., 1996).


The importance of phosphate supply in regulating the growth of cyanobacteria is widely recognized, especially in aquatic environments. Increased phosphate levels together with favourable weather conditions, for example water surface temperatures over 20 °C, often result in the development of widespread cyanobacterial blooms. The relationship between the available amounts of phosphate and the production of EPS is not straightforward, as the overall effect might be dependent on a set of interlinked variables such as the amount of phosphate, nitrate and sulphate (Grillo & Gibson, 1979). In most cases, phosphate starvation or low levels of phosphate induced an increase in EPS production (De Philippis et al., 1993; Roux, 1996; Nicolaus et al., 1999; Huang et al., 2007); however, in C. capsulata, the absence of phosphate had no significant effect (De Philippis et al., 1991), and in Anabaena spp. and Phormidium sp., it significantly decreased EPS production (Nicolaus et al., 1999). Generally, an increase in phosphate concentration in the growth medium has little effect on the amount of exopolymers.


Cyanobacterial EPS contain sulphate groups, a unique feature among bacteria and shared by the EPS produced by archaea and eukaryotes (Sutherland, 1994; De Philippis et al., 1998; De Philippis & Vincenzini, 2003; Micheletti et al., 2008b). It has been reported that sulphur limitation has a dramatic impact on the cells, resulting in morphological and physiological changes similar to those due to nitrogen limitation (Wanner et al., 1986). In Gloeothece sp. PCC 6909, sulphur starvation caused significant morphological alterations in the cells, such as the synthesis of a structureless sheath, the accumulation of cyanophycin, polyhydroxybutyrate and glycogen granules and the disintegration of thylakoid membranes. Most of these changes were reversed by the addition of sulphate to the culture (Ortega-Calvo & Stal, 1994; Ariño et al., 1995).

Salt (NaCl)

The acquisition of salt tolerance in some cyanobacteria living in extreme environments induces various structural and metabolic changes, including a decrease in respiration and an increase in the production of some carbohydrates, notably sucrose, which functions as an osmotic solute protecting membranes from desiccation (Chen et al., 2006). Generally, under salt stress (about 0.5 M), cyanobacteria also produce larger amounts of EPS (Table 5). It has been postulated that the increased export of EPS can have a function equivalent to that of sucrose, i.e. improving salt tolerance and carbohydrate metabolism (Chen et al., 2003). However, some exceptions are reported, in which an increase in NaCl concentration did not affect or even lowered the EPS productivity. In C. capsulata (De Philippis et al., 1991) and Cyanothece sp. 16Som2 (De Philippis & Vincenzini, 1998), the presence of a thick and firmly attached capsule probably provided enough protection against osmotic shocks, and in Synechococcus sp., EPS production increased only in the stationary phase, possibly because a nutrient limitation is necessary for the activation of EPS production (Roux, 1996). In Anabaena sp. ATCC 33047 growing under diazotrophic conditions, EPS production was enhanced only under conditions in which the nitrogenase activity and phycobiliprotein content were low, and production decreased in the presence of higher NaCl concentrations. However, the authors did not provide any explanation for this behaviour (Moreno et al., 1998). In the halophilic cyanobacterium Aphanothece halophytica GR02 grown in the presence of various NaCl concentrations, a variation in the relative amounts of rhamnose and galactose, two of the seven monosaccharides constituting the RPS, was observed (Li et al., 2001).


Aeration seems to be vital for increasing the production of EPS by cyanobacteria, with the few studies available reporting that EPS production reached a maximum with continuous aeration (Moreno et al., 1998; Nicolaus et al., 1999; Su et al., 2007). A possible explanation is that the increase in culture turbulence may facilitate the release of the polysaccharides from the cell surface, thus stimulating the synthesis of new exopolysaccharides. However, it is also possible that the higher turbulence due to the aeration provides a better stirring of the viscous cultures, which may increase the light penetration and consequently may induce a higher biosynthetic activity of the cells.


The majority of the studies dealing with the production of EPS in cyanobacteria use the optimal growth temperature for the organism under investigation and, again, the limited data available indicate that the effect of the temperature variation is strain dependent. For Anabaena sp. ATCC 33047, an increase in the temperature (from 30/35 to 40/45 °C) led to a noticeable increase in the production of the EPS (Moreno et al., 1998), probably because at higher temperatures, the time required to reach the onset of stationary phase was shorter than that required at 30/35 °C. In contrast, the temperature increase (from 30 to 35 °C) did not affect the EPS productivity in Nostoc sp. PCC 7936 (Otero & Vincenzini, 2004), and temperatures >30 °C even caused a small decrease in EPS production in Spirulina sp. (Nicolaus et al., 1999).


The synthesis and release of EPS are particularly light dependent, even though different light regimens (continuous light and light–dark cycles) do not seem to have a significant effect on the quality of the polymer, i.e. monosaccharidic composition and relative proportions of the sugar units (Vincenzini et al., 1993; De Philippis & Vincenzini, 1998). However, generally, EPS production is enhanced by continuous light and high light intensities (up to 400 μmol photons m−2 s−1), but it is important to consider both the culture volume and the geometry of the culture flasks/bioreactors when adjusting the light intensity (see, e.g. Fischer et al., 1997). Moreover, it was demonstrated that certain wavelengths influence EPS production; notably, in the heterocystous N. commune, UV-B irradiation stimulates extracellular glycan production as well as induces the synthesis of photoprotective pigments (Ehling-Schulz et al., 1997).

Other factors

Many other factors that can influence EPS production, notably pH, dilution rate, growth phase, presence/absence of magnesium, calcium, potassium and heavy metals, as well as the addition of glycoxylate, acetate, valerate, glucose, citrate and EDTA have been sporadically studied (Li et al., 2002), but not consistently evaluated.

In summary, although the key factors controlling the production of the cyanobacterial EPS have been identified, comprehensive strain-specific studies taking into account the interaction between the variables to understand the system response to changes, are still missing. This requires a better knowledge of the genes and metabolic pathways involved in the production of EPS in cyanobacteria.

Genes and biosynthetic pathways related to the production of EPS

Over the past decade, several studies have been initiated to try to understand the genetics and biochemistry of EPS biosynthesis in bacteria (Van Kranenburg et al., 1999; De Vuyst et al., 2001; Jolly & Stingele, 2001; Laws et al., 2001; Sutherland, 2001; Welman & Maddox, 2003; Whitfield, 2006). However, cyanobacteria have not been thoroughly examined and, consequently, the information available is extremely limited (Yoshimura et al., 2007). Studies performed in both Gram-negative and Gram-positive bacteria revealed that the EPS biosynthetic pathways are very complex, including, besides the enzymes directly involved in the EPS synthesis, enzymes engaged in the formation of the cell wall polysaccharides and lipopolysaccharides (Mozzi et al., 2003). However, the mechanisms involved in the synthesis of EPS are relatively conserved throughout bacteria. Typically, this process comprises four distinct steps occurring in different cellular compartments: (1) the activation of the monosaccharides and conversion into sugar nucleotides in the cytoplasm, (2) the assembly of the repeating units by sequential addition of sugars onto a lipid carrier by glycosyltransferases at the plasma membrane, (3) the polymerization of the repeating units at the periplasmic face of the plasma membrane and (4) the export of the polymer to the cell surface (Yamazaki et al., 1996; De Vuyst & Degeest, 1999; Kleerebezem et al., 1999; Whitfield & Roberts, 1999; De Vuyst et al., 2001; Jolly & Stingele, 2001; Sutherland, 2001). A schematic representation is depicted in Fig. 1. The sugar activation/modification enzymes and the glycosyltransferases are strain dependent, whereas the proteins involved in the polymerization, chain length control and export are conserved among bacteria. Some of these conserved proteins, as well as their interactions, are highlighted in Fig. 1, and their putative roles are discussed below.

Figure 1.

 Sequence and compartmentalization of the putative biosynthetic pathway, polymerization and export of EPS in cyanobacteria, based on the information gathered for other bacteria as well as the genes present in the available cyanobacterial genomes. Some of the proteins involved in the polymerization, chain length control and export are highlighted along with the interactions among them (dashed line arrows).

The genes related to the production of surface polysaccharides can be divided into three classes: (1) those encoding the enzymes involved in the biosynthetic pathways of nucleotide sugars, or other components, needed for polysaccharide synthesis and not otherwise available in the cells; (2) those coding for the glycosyltransferases; and (3) those required for the oligosaccharide or polysaccharide processing (Reeves et al., 1996).

The first class is a vast and diverse group of genes not specific for EPS biosynthesis, given that sugar nucleotides are needed for the synthesis of a range of polysaccharides, and this group, therefore, will not be extensively discussed in this work. Among these genes are rfbABCD, also frequently called rml genes (Reeves et al., 1996), which encode proteins involved in the biosynthesis of l-rhamnose. l-Rhamnose is a 6-deoxyhexose commonly present in bacteria, but only as a component of surface polysaccharides (Li & Reeves, 2000). Indeed, dTDP-rhamnose is commonly found in the EPS of Gram-negative and Gram-positive bacteria (Li & Reeves, 2000) and is a key constituent of the O-antigens of lipopolysaccharides of Gram-negative bacteria (Reeves, 1993). Furthermore, rhamnose is one of the sugars frequently found in the cyanobacterial EPS, and the proteins encoded by the rfb genes are listed in CyanoBase ( as involved in the assembly of cyanobacterial surface polysaccharides. However, the participation of these proteins in the biosynthesis of both lipopolysaccharides and EPS makes it very difficult to determine their specific role. Moreover, the presence of several acidic or neutral monosaccharides in cyanobacterial EPS indicates that their biosynthetic pathway may be even more complex (Sutherland, 2001; Li et al., 2002).

Glycosyltransferases are key enzymes for the biosynthesis of the EPS repeating unit, catalyzing the transfer of the sugar nucleotides from activated donor molecules to specific acceptor molecules – most probably a lipid carrier – in the plasma membrane. A large number of genes encoding glycosyltransferases have been identified, given the structural diversity of the bacterial extracellular polysaccharides, and consequently the number of possible linkages. The diverse function of the transferases, which in addition are strain specific, is reflected in the heterogeneity of their DNA sequences (Reeves et al., 1996; De Vuyst et al., 2001; Jolly & Stingele, 2001; Samuel & Reeves, 2003). In silico analysis of the cyanobacterial genomes revealed the presence of numerous genes putatively encoding glycosyltransferases; however, the enzymes have not been biochemically characterized, which makes it impossible to assign their function to the synthesis of EPS.

In bacteria, the genes encoding the proteins responsible for polymer extension and processing are usually clustered and organized in a similar way (De Vuyst & Degeest, 1999), often constituting long operons. Within these clusters, three different regions can be discerned: a central region constituted by the genes encoding the glycosyltransferases, flanked by two regions comprising the genes encoding enzymes involved in the polysaccharide polymerization, chain length control and export (De Vuyst & Degeest, 1999). The nomenclature of the latter genes is very diverse, for example being named eps and cps for lactic acid bacteria, wz_ and kps for Escherichia coli, exo for Sinorhizobium meliloti and gum for X. campestris (De Vuyst & Degeest, 1999; De Vuyst et al., 2001; Jolly & Stingele, 2001; Sutherland, 2001; Welman & Maddox, 2003; Whitfield & Paiment, 2003; Whitfield, 2006). Recently, a region containing 18 ORFs putatively involved in polysaccharide biosynthesis was identified for the cyanobacterium Anabaena sp. PCC 7120 (Yoshimura et al., 2007).

Despite the variety of bacterial exopolysaccharides, bacteria use a limited repertoire of assembly and secretion strategies, which are represented in E. coli (Whitfield & Roberts, 1999; Whitfield & Paiment, 2003; Whitfield, 2006). For this organism, two models have been proposed for the biosynthesis and assembly of the different types of capsules based on genetic and biochemical criteria, with the one proposed for the capsules of groups 1 and 4 being the most common among Gram-negative bacteria (Rahn et al., 1999; Whitfield, 2006; Whitfield & Larue, 2008). This mechanism is Wzy-dependent, in contrast to the mechanism for groups 2 and 3 capsules, which are assembled via ABC-transporter-dependent pathways (Whitfield, 2006). Using the E. coli Wzy-dependent model, together with the information derived from Anabaena sp. PCC 7120 (Yoshimura et al., 2007) and the cyanobacterial genome sequences, a putative mechanism was put forward for cyanobacteria (Fig. 2). This is a hypothetical working model, on which further studies aiming to elucidate the mechanisms involved in the production of cyanobacterial EPS can be based. Assuming that a lipid carrier (in most of the Gram-negative bacteria an undecaprenol diphosphate – see Skorupska et al., 2006) is also present in cyanobacteria, the glycosyltransferases will pass the sugar nucleotides to this acceptor, where repeating units are assembled. This step takes place at the interface between the cytoplasm and the plasma membrane. The newly synthesized lipid-linked repeating units are then flipped across the membrane in a process requiring Wzx, an integral plasma membrane protein. This provides the substrate for the blockwise polymerization of the repeating units that takes place at the periplasmic face of the membrane, a step carried out by the Wzy protein. The polymerization also requires the auxiliary protein Wzc to act at the interface between the membrane and the periplasmic space (Skorupska et al., 2006), probably for the control of the chain length of the growing polymer. Transphosphorylation of Wzc and its dephosphorylation by Wzb is required to regulate the polysaccharide polymerization and export. The translocation process is mediated by the outer-membrane auxiliary protein Wza, which forms a channel, allowing the externalization of the growing polysaccharide to the cell surface. The translocation may require the physical association of proteins located in both membranes, notably Wzc and Wza (Whitfield & Paiment, 2003; Skorupska et al., 2006; Whitfield, 2006; Collins & Derrick, 2007).

Figure 2.

 Proposed model for the assembly and export of cyanobacterial EPS based on the information gathered both for other bacteria and genes found in cyanobacterial genomes. (1) Glycosyltransferases transfer the nucleotide sugars onto a putative lipid carrier, and the lipid-linked repeated units are assembled at the interface between the cytoplasm and the plasma membrane. (2) Newly synthesized units are ‘flipped’ across the membrane in a process requiring the integral membrane protein Wzx. Subsequently, Wzy assembles the polysaccharide by addition of new repeating units to the growing polysaccharide chain. (3) The polymer is translocated in a process requiring the Wzc and Wzb proteins. In the final stage, the carbohydrate polymer is translocated across the outer membrane through the outer-membrane lipoprotein Wza (adapted from Whitfield, 2006). OM, outer membrane; PM, plasma membrane.

An in silico analysis of the cyanobacterial genomes revealed that the genes putatively involved in the production of exopolysaccharides are sometimes clustered, present in different regions of the genome, and often occur in multiple copies. This last feature is not common in E. coli, Klebsiella pneumoniae and lactic acid bacteria, where the genes are frequently clustered and transcribed as one or two operons (Roberts, 1996; De Vuyst & Degeest, 1999; De Vuyst et al., 2001; Jolly & Stingele, 2001; Whitfield & Paiment, 2003; Whitfield, 2006). Examples of the physical organization of wz_ genes in three morphologically distinct types of cyanobacteria are depicted in Fig. 3. For the unicellular Cyanothece sp. only two copies of each gene were found, with the exception of wzx, which appears to be single (its sequence may not be complete). It was not possible to determine the relative position of these genes for this organism as the genome annotation process is still in progress. In general, as the complexity of the organism/size of the genome increases, more copies of the genes putatively involved in the production of the EPS are found, as can be observed for Lyngbya sp. and Nostoc punctiforme. In the last case, one needs to consider the presence of heterocysts that also have a polysaccharidic layer in their envelope. However, in the filamentous strains, a genome region containing all wz_, except wzb, could be identified, but it remains to be shown whether these genes are indeed specifically related to EPS production, and whether they constitute a transcriptional unit. Only by construction of deletion mutants will it be possible to understand the function of each of the proteins encoded by these ORFs and to start to unveil the biosynthetic pathways of cyanobacterial EPS.

Figure 3.

 Physical map of the putative genes involved in the polymerization, chain length control and export of EPS in the three morphologically distinct types of cyanobacteria. The deduced protein sequences encoded by these genes were submitted to an in silico analysis to identify the conserved motifs of interest. This analysis was performed using the following bioinformatic tools: blastp, cdart (at NCBI –, and smart (at EMBL – In general, several copies of a specific gene could be identified in a single cyanobacterial strain. In a given organism, the copy that has the highest probability to be related to EPS production is designated by subscript 1 (taking into account both the percentage of identity with the corresponding sequences in other organisms and the position of the gene in relation to others involved in the same process); the other copies are numbered subsequently. In Lyngbya sp. and Nostoc punctiforme, the genome region containing all wz_, except wzb, is underlined with a dashed line. ? indicates that the gene wzy2 from Lyngbya is the one with the lowest homology to the available wzy sequences. Genbank accession numbers: AAXW00000000 (Cyanothece sp. CCY 0110), AAVU00000000 (Lyngbya sp. PCC 8106) and CP001037 (N. punctiforme ATCC 29133).

The synthesis and secretion of EPS in cyanobacteria probably follow the pathways previously described for other bacteria. However, as a consequence of the cyanobacteria's ability to perform oxygenic photosynthesis and the unique characteristics of their EPS, some differences are expected. The production of exopolysaccharides is intimately dependent on the balance between the catabolic pathways of sugar degradation and the anabolic pathways of sugar nucleotide synthesis. This balance is certainly different in cyanobacteria compared with heterotrophic bacteria. Moreover, the presence of a higher number of different sugars in cyanobacterial EPS suggests that the synthesis of the sugar nucleotides is more complex, involving a higher number of different enzymatic reactions. In addition, although several genes encoding proteins putatively involved in the Wzy-dependent mechanism of EPS polymerization and export were identified in cyanobacterial genomes, their physical organization differs from what is observed in other microorganisms, suggesting that in cyanobacteria, this mechanism may be under a different type of regulation.

Concluding remarks

As discussed previously by De Philippis & Vincenzini (1998), the data on the chemical composition and on the rheological properties of the cyanobacterial EPS are not always comparable due to the different hydrolytic procedures and analytical methods used. Therefore, some of the results reported in the literature were not included in this review as they were not consistent with the majority of the data available.

The information gathered strongly underlines the complexity of both the chemical features of the cyanobacterial EPS and their putative biosynthetic pathways. As a result, it is not surprising that the data available on the structures of these macromolecules are still scarce and little is known about the genes encoding the proteins involved in their synthesis. Consequently, it is important to generate knowledge to unveil the pathways utilized by cyanobacteria for the synthesis of these biopolymers, which not only play a decisive ecological role, allowing these organisms to survive in adverse environmental conditions, but also have a high potential for biotechnological applications. The identification of the genes involved in the biosynthesis of EPS would also offer the possibility to investigate (1) the factors regulating the expression of these genes and (2) the possible genetic modification that could be introduced. This will make it possible to maximize the production of the polymer, as well as to introduce specific alterations in the composition/structure, producing polymers more suitable for specific applications. The construction of deletion mutants will help to define the role of each gene product and to clarify the function of EPS in natural habitats.


This work was supported by Fundação Calouste Gulbenkian: Programa Ambiente e Saúde, Proc. No. 76910; FCT (SFRH/BD/22733/2005 and SFRH/BPD/37045/2007), POCI 2010 (III Quadro Comunitário de Apoio) and by Acordo de Cooperação Científica e Tecnológica GRICES/CNR, Proc. 4.1.1., 2007.